On 10/30/2012 10:03 AM, Viresh Kumar wrote:
On 30 October 2012 14:30, Daniel Lezcano daniel.lezcano@linaro.org wrote:
It is very curious if we disable all the configs option, a slave creation raise a BUG (cf. dsa_slave_create). IIUC, booting with NET_DSA enabled and none of the NET_DSA_TAG* enabled will raise a BUG in the probe function, right ?
Maybe we should force at least one config when none are set ?
I thought of it earlier. But found the one which i posted better. As we will not compile DSA stuff now without these tags.
Well, it is the same here, no ? Except, it is up to the user to disable the option.
-- viresh
diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index 274791c..86326e3 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -1,8 +1,9 @@ -config NET_DSA +menuconfig NET_DSA tristate "Distributed Switch Architecture support" default n depends on EXPERIMENTAL && NETDEVICES && !S390 select PHYLIB
select NET_DSA_TAG_DSA if (!NET_DSA_TAG_EDSA &&!NET_DSA_TAG_TRAILER) ---help--- This allows you to use hardware switch chips that use the Distributed Switch Architecture. @@ -12,15 +13,15 @@ if NET_DSA
# tagging formats config NET_DSA_TAG_DSA
bool
bool "tag dsa" default nconfig NET_DSA_TAG_EDSA
bool
bool "tag edsa" default nconfig NET_DSA_TAG_TRAILER
bool
bool "tag trailer" default nendif
-- http://www.linaro.org/ Linaro.org │ Open source software for ARM SoCs
Follow Linaro: http://www.facebook.com/pages/Linaro Facebook | http://twitter.com/#!/linaroorg Twitter | http://www.linaro.org/linaro-blog/ Blog